The pathophysiologic model of heart failure with preserved ejection fraction (HFpEF) has evolved over the past several years with evidence suggesting a role of metabolic impairment in disease pathogenesis. Due to the increasing frequency of and therapeutic challenges associated with HFpEF, there is a need to improve our understanding of subclinical HFpEF correlates to prevent disease manifestation. Diastolic dysfunction is one key defining feature on the causal disease pathway of HFpEF, yet the metabolic profile and nutritional risk factors associated with this subclinical phenotype remain ill-defined, especially in diverse populations. To address this knowledge gap, we propose the first epidemiological study aimed at identifying novel metabolites and dietary patterns associated with diastolic dysfunction. We will leverage untargeted metabolomics data, along with longitudinal measures of both diastolic function and dietary protein intake in up to 912 Bogalusa Heart Study (BHS) participants. The BHS is a NIH-funded longitudinal study that examines cardiovascular health across the lifespan, providing an ideal setting for the proposed work. Left ventricular diastolic function will be assessed through seven key 2-dimensional and Doppler echocardiography parameters: mitral annular velocity, peak early filling velocity, peak velocity caused by atrial contraction, isovolumic relaxation time, deceleration time, left atrial maximal volume index, and peak tricuspid regurgitation velocity. Likewise, given the blood pressure lowering properties of dietary protein, a key mechanism that may help preserve diastolic function, we will characterize dietary patterns by focusing on different sources of protein consumption. In Aim 1, we will take advantage of untargeted serum metabolome profiles of BHS participants, derived using ultrahigh performance liquid chromatography-tandem mass spectroscopy, to examine novel associations of metabolites and metabolite pathways with diastolic function in adults. Multivariable-adjusted methods and network-based statistical approaches will be used to test associations of metabolites and metabolite pathways with diastolic function in combined and race-stratified analyses. To validate results, novel BHS findings will be tested for significance among 1,000 adults of the Multi-Ethnic Study of Atherosclerosis. Dietary patterns in childhood and adolescence will then be characterized by: 1) calculating mean intakes of six major protein sources, meat, poultry, dairy, fish, grains, and vegetables, across the two time-points; and 2) estimating the change in intake of major protein sources between the two time-points. We will test whether these specific protein consumption patterns associate with diastolic function, controlling for covariables collected at the time of dietary assessment (Aim 2). In Aim 3, we will conduct integrative studies to identify dietary protein-derived metabolites and detect biological pathways underlying diastolic dysfunction. In total, the proposed research may provide important information on biological mechanisms underlying diastolic dysfunction and guide the development of novel biomarker or dietary strategies to prevent HFpEF.